Rebuilding Microsoft Exchange
By: Ryan Skarra-Gallagher
Senior Systems Engineer
Earlier this year Coyote Creek was contacted by a mid-sized East Coast financial services provider that’s been running an on-premise installation of Microsoft Exchange 2013. Something had gone wrong with their Exchange system, and now Microsoft had told them that the best solution was rebuilding Microsoft Exchange completely. They called us to help them do so.
Problem: A repair reinstall broke the offline address book
The company’s problems started when their single Exchange server failed. Attempts to recover this failed Exchange server were unsuccessful. It was determined that the only fix was to do a repair installation of Exchange on a new server with the same name. With the Microsoft technicians on the phone guiding them, they did the reinstall.
When you do a repair reinstall there’s a requirement that you use the same server name and drive as before. This is so that the system can access important information that is stored in Active Directory. For example, during the reinstall the system will pull configuration data out of Active Directory, and reuse the pointers from the previous installation.
Unfortunately, in doing this repair reinstall, the company’s IT staff did something wrong. Their error left some of the pointers pointing at where information used to be in the old server, instead of where it was now residing in the new server. As a result, the offline address book broke. When the offline address book went to download, it was looking at a path that no longer existed, so it would completely fail.
Solution: Rebuilding Microsoft Exchange – and make it better than before
Once they realized that the offline address book was dead, the company opened a case with Microsoft. After troubleshooting the problem, the Microsoft technicians informed them that to fix the problem they needed to go through with rebuilding Microsoft Exchange.
As long as they were rebuilding Microsoft Exchange, the company decided to take this opportunity to redesign and improve it. Their old system only had one server. Now they wanted to add high availability, to avoid this single point of failure risk.
We met with them, discussed their needs, and then designed a resilient architecture for them that included high availability. Their new system includes:
- Two multi-role Exchange 2013 servers, each with both the mailbox and client access roles
- A virtual load balancer to provide high availability for the client access role
- A DAG (Database Availability Group) setup to provide high availability for the mailbox role
Result: High availability plus a functioning offline address book
With the new system there are copies of the mailboxes and databases on both servers, with things load balanced between them. This gives the system less load and twice as much horsepower. Plus, our client can sleep better at night knowing that their vital Exchange system has redundancy. If something goes wrong with one server, there’s a backup server in place to handle all functionality immediately. If they need to do maintenance on a server, they can do so without bringing email down.
Of course, the offline address book works perfectly in their new Exchange installation as well.
Now that they have a recommended architecture in place, the company has also signed up for Coyote Creek’s Remote Monitoring and Management services. Which means that we’re monitoring these servers 24/7, providing patching, maintenance and triage for them.
As Microsoft Gold Certified Partners with over 20 years of experience successfully helping enterprise-level organizations make the most of Microsoft technology, we have the expertise to help you with any Exchange-related issue you may have. We can design and install a new system, upgrade an existing system, monitor and manage your existing environment, or troubleshoot those nagging, intermittent or mysterious problems that are driving your IT team crazy. Just give us a call and let us know what you need.